search

IntelLabs / coach

Reinforcement Learning Coach by Intel AI Lab enables easy experimentation with state of the art Reinforcement Learning algorithms
https://intellabs.github.io/coach/
Apache License 2.0
2.32k stars 460 forks source link

Tf2 migration #393

Closed dandanelbaz closed 4 years ago

dandanelbaz commented 4 years ago

WIP: Starting migration to TensorFlow 2. Tested only for Cart-pole DQN. Optimizer, loss, and DNN model are implemented in different classes and instantiated inside trainer class. Head and loss are separated. The following UML class diagram provides a high level description

image

dandanelbaz commented 4 years ago

The only way I managed to stop sending alerts to Neta is by closing the PR